Using Single-Channel Blind Deconvolution to Choose the Most Realistic Pharmacokinetic Model in Dynamic Contrast-Enhanced MR Imaging

摘要: In dynamic contrast-enhanced magnetic resonance imaging, there has been no consensus in the choice of pharmacokinetic model. this paper, a new approach for assessment most realistic model given tissue is presented. Non-blind and single-channel blind deconvolution algorithms were used quantitative imaging mouse masseter muscle to compare realism two different models residue function. The first was adiabatic approximation homogeneity (aaJW) second, two-compartment exchange (2CXM). Normals mice treated with substance C48/80 studied. increases both blood flow contrast leakage substantially. obtained accuracy evaluated models. addition, arterial input functions (aifs) estimated using compared corresponding observed aifs. hypothesis that function leads best fits. non-blind did not result any clear answer. For deconvolution, aifs aaJW very similar aifs, clearly more so than 2CXM Also, tracer time sequences accurate describing passage through microvascular bed single could be assessed deconvolution.